Diabetes, especially when there is poor control of blood sugar levels, leads, over time to a series of debilitating complications. These diseases of blood vessels affect the eyes, kidneys, heart, nerves. They are a major cause of blindness, renal failure, stroke and heart attack. One main culprit is hyperglycemia or high blood sugar. And the links are shown here

Diabetes and its complications. The evidence

Diabetes mellitus, a condition characterized mainly by a quantitative deficiency in insulin secretion or a resistance to insulin action, is estimated to afflict 5-7% of the population. Microangiopathy, the microvessel disease in diabetes, includes retinopathy, nephropathy, and neuropathy and in type 1 patients the first signs of these complications may develop even in adolescence, particularly if insulin treatment has been inadequate. Similar complications occur later in life in type 2 patients and are frequently present at the time of diagnosis.The precise mechanisms by which diabetic microangiopathy develops are not fully understood, but a consensus is emerging pointing to a terrain of genetic influences onto which metabolic and hemodynamic derangements are superimposed Apart from the key hemodynamic changes intervening in many tissue targets of diabetic complications, there is sound clinical and epidemiological evidence that links hyperglycemia to vascular complications.

Two controlled clinical trials stand out:

a) The Diabetes Control and Complications Trial (DCCT).

b) The United Kingdom Prospective Diabetes Study (UKPDS).

The Diabetes Control and Complications Trial (DCCT).

The DCCT was designed to definitively answer the question of the association between hyperglycemia and vascular complications in a cohort large enough to permit incontestable conclusions (2). The DCCT evaluated intensive insulin replacement and self-monitoring of blood glucose, and utilized glycated hemoglobin assays to measure glycemic control over long periods of time.

The data published in 1993 showed that there was a 60% reduction in risk between the intensive treatment group and the standard treatment group in diabetic nephropathy, retinopathy and neuropathy. The outcome showed that reduction of HbA1c from levels of approximately 9% to approximately 7% reduced the progression and/or development of all microvascular complications. This change was due for the most part to the effect of therapy on glycemic control and, to some extent to the methods employed to achieve that control. All categories of patients benefited from intensive therapy, irrespective of age, sex, or duration of diabetes.

B) The United Kingdom Prospective Diabetes Study (UKPDS) is a randomized trial of intensive treatment of type 2 diabetics followed for several years. The study recruited over 5000 patients with newly diagnosed type 2 diabetes in 23 centers in the U.K. between 1977 and 1991. The UKPDS started by analyzing the value of various strategies (diet and several oral hypoglycemic agents) to achieve tight blood glucose control compared with looser control.

The UK study has provided answers to a range of important questions that have plagued diabetes researchers and physicians for decades. Tightly controlling blood glucose concentration reduced the risk of complications in type 2 diabetes. The overall microvascular complications rate was decreased by 25% in patients receiving intensive therapy versus conventional therapy (13-15). Confirming the DCCT data the UKPDS showed a continuous relationship between the risk of microvascular complications and glycemia. For every percentage point decrease in HbA1c there was a 35% reduction in the risk of microvascular complications. As far as reducing the risk of microvascular complications is concerned, sulphonylureas and insulin produced equally good results.

Implications in clinical practice

In its position statement, the ADA expresses agreement with the design, protocols and randomization of patients in this study. Clinical and laboratory tests were done by recognized methodologies, and all end points were adequately documented. Moreover the committee is confident that the results should apply to the U.S. population of men and women with type 2 diabetes.

Thus, the hypothesis that it is glucose itself that is toxic in type 2 diabetes is confirmed, in line with the findings of the DCCT for type 1 diabetes: the controversy should now end and we must unravel the mechanisms of this effect in order to better approach it therapeutically. The achievement of tight blood glucose control in type 2 diabetes is feasible and should become the standard of care. The combination of pharmacologic agents should be based on the individual evaluation of each patient. Notwithstanding the failure of diet therapy alone, diet remains an adjunct to pharmacologic therapy. It must be acknowledged that the ability to prevent or at least retard these complications may be made easier in the near future by the recently approved hypoglycemic agents that were not available to the UKPDS.

How high blood sugars and complications are linked?

It is important to point out that no consensual framework has been found which encompasses all that is known about the link between hyperglycemia and complications. There are several equally defensible hypotheses on the roots of complications.Three of the favored pathways that are being investigated and that potentially explain the mechanisms by which high glucose levels can result in vascular damage are:

A. The sorbitol theory,

B. Modification of protein kinase C activity, and

C. The glycation hypothesis, which will be the main focus

It cannot be overemphasized that oxidative stress is generated in all these three pathways as well as in several others, and its importance will be outlined when considering each of them.

The sorbitol or polyol pathway

The sorbitol hypothesis was proposed almost three decades ago. As depicted here, high glucose concentrations in non insulin-dependent tissues may follow the pathway of aldose reductase. Since this is a high Km enzyme its activity is very low when glucose concentrations are normal. Sorbitol is the product of this reaction and NADPH is used as a cofactor. In a variety of animal models of hyperglycemia, increases in sorbitol formed through this reaction lead to altered cellular-energy metabolism, cell-membrane integrity, and other functions. This is one possible biochemical mechanism by which hyperglycemia could impair the function and structure of the cells affected by diabetic complications

Protein kinase C activity. Another role of hyperglycemia appears to be the modification of protein kinase C (PKC) activity by hyperglycemia-induced increases in diacylglycerol (DAG), partly due to de novo synthesis. This chain of events should increase PKC activity. However, in tissues where aldose reductase levels are high, the opposite seems to be true, as depicted. Decreased levels of myoinositol, probably shifted outside the cell when sorbitol levels increase, result in modification of phospho-inositol and diacylglycerol (DAG) metabolism which in turn affect PKC function. PKC regulates various vascular functions by modulating enzymatic activities, such as cytosolic phospholipase A2 and Na+/K+ATPase, or gene expressions of extracellular matrix components and contractile proteins. When PKC activity is poorly regulated some of the resulting vascular abnormalities include changes in retinal and renal blood flow, contractility, permeability, and cell proliferation .

What's glycation?

Sweet and sour, or the age of AGE

In the glycation reaction, first discovered by the French chemist L. Maillard in 1912 while studying foods, sugars react non-enzymatically with a wide range of proteins to form early glycation (Amadori or fructosamine) products. In humans, this process was first demonstrated for hemoglobin but almost any protein can be affected. Clinically, the measurement of the glycated form of hemoglobin, HbA1c, has revolutionized the monitoring and the study of diabetic patients. Fructosamine (to be chemically correct, fructosamino-protein adduct) is the common name given to any glycated plasma protein in this first stage. Measurement of glycated plasma proteins (usually called ‘the fructosamine assay’) is used as a tool for monitoring glycemic control over a 3 week period .

The afore-mentioned reactions are considered ‘early glycation’ and they are by no means the end of the reaction cascade. In a second phase of the glycation pathway, a complex series of rearrangements and oxidative reactions leads to the formation of multiple, very reactive species, collectively named advanced glycation end products or AGE-products , some of which are shown in Figure 4. Incidentally, a similar reaction, even though more complete and produced by harsher conditions, occurs between sugars and proteins in foods, and the final result is what we see in bread or piecrusts, for instance. The Maillard reaction also plays a part in the generation of brownish pigments in beer and cola drinks.

As stated above, the reactive dicarbonyl intermediates, formed from Amadori products or from sugars, react with protein amino groups to form a variety of AGEs. AGE-products accumulate in vivo on vascular wall collagen and basement membranes as a function of age and levels of glycemia. They are capable of producing cross-linking of proteins and have been shown to display diverse biological activities .

Inherited differences in the ability to detoxify AGE intermediates might be one of the genetic factors responsible for the clinically observed large variability that the impact of a given level of glycemia has on diabetic complications.

Whatever the case is, AGE molecules are found in plasma, cells, and tissues and accumulate in the arterial wall, the kidney mesangium, and glomerular and other basement membranes.

What's glycation?

Sweet and sour, or the age of AGE

Outline of the reactions, sugars attach to proteins, a 2 phase series of reactions ensues

But we said diabetes leads to microangioptahy. How do these AGEs affect microvessels?

Considerable amounts of data on this issue have been accumulating during the last few years; we will select and provide an outline of what we consider most clinically significant.

Direct effects of AGEs on proteins. AGEs and extracellular matrix (ECM). AGE formation modifies the functional properties of different key extracellular matrix molecules. In collagen (the most abundant protein in the body) AGEs form covalent, intermolecular bonds . As depicted luminal narrowing, a major feature in diabetic vessels may arise in part from accumulation in the subendothelium of plasma proteins such as albumin, low-density lipoprotein (LDL) and immunoglobulin G (IgG). They may get trapped in basement membranes by covalently cross-linking to AGEs on collagen. It is well known that the main features in diabetic glomerulopathy are proteinuria, mesangial expansion, and focal sclerosis. How do AGE contribute to this? AGE formation on laminin (a key structural protein of the ECM) causes reduction in polymer self-assembly and decreased binding of the other major components of the molecular scaffolding of the basement membrane, namely type IV collagen and heparan sulfate proteoglycan . Heparan sulfate proteoglycan (HSPG) which provides the negative charge of glomerular basement membrane (GBM) is per se the key factor impairing the leaking of plasma proteins and the resultant proteinuria

AGE formation modifies the functional properties of different key extracellular matrix molecules.

In collagen (the most abundant protein in the body) AGEs form covalent, intermolecular bonds . As depicted luminal narrowing, a major feature in diabetic vessels may arise in part from accumulation in the subendothelium of plasma proteins such as albumin, low-density lipoprotein (LDL) and immunoglobulin G (IgG). They may get trapped in basement membranes by covalently cross-linking to AGEs on collagen.

Diabetes-induced loss of matrix-bound heparan sulfate proteoglycan, secondary to AGE modification of glomerular basement membrane proteins , could prompt protein leaking and stimulate a compensatory overproduction of other matrix components in the vessel wall.

This provides a strong molecular support to diabetic Kimmelstiel-Wilson nephropathy . On the other hand, these AGE-induced abnormalities alter the structure and function of microvessels other than the renal microcirculation.

Receptor-Mediated effects

Mononuclear cells. Monocytes and macrophages were first shown to bear specific receptors for AGEs (RAGE). As illustrated on AGE proteins binding to these receptors stimulate macrophage production of interleukin-1, insulin-like growth factor I, tumor necrosis factor alfa, and granulocyte/macrophage colony-stimulating factor at levels that have been shown to increase glomerular synthesis of type IV collagen and to stimulate proliferation of both arterial smooth muscle cells and macrophages .

Endothelium. As shown in diagram form , reactive oxygen species (ROS) are generated after AGE binding to endothelial cells where they activate the free radical-sensitive transcription factor NFkB, a multi-faceted coordinator of numerous "response-to-injury" genes . These AGE-induced changes are involved in the modification of thrombomodulin and tissue factor production. These alterations prompt two cumulative pro-coagulant changes in the endothelial membrane . Concurrently, these AGE-induced alterations in endothelial cell function favor thrombus formation at sites of extracellular AGE accumulation .

The colocalization of receptors and AGEs at the microvascular sites of injury suggests that their interaction may play a significant role in the pathogenesis of diabetic vascular lesions.

Several receptors for AGEs have been characterized

Main reactions triggered by their occupancy

Mesangial cells in kidney glomeruli.

AGE receptors have also been described on glomerular mesangial cells AGE protein binding to their receptors on mesangial cells stimulates platelet-derived growth factor secretion, which in turn mediates mesangial expansion . In vivo, chronic administration of AGEs to otherwise healthy and euglycemic rats leads to focal glomerulosclerosis, mesangial expansion, and proteinuria, the hallmarks of diabetic microangiopathy.

Not only blood sugar...and not only diabetes

In the past few years it has been shown that there are other reactive molecules in our bloodstream. Low molecular weight peptides containing AGE circulate at increased levels in plasma from diabetic and kidney failure patients. These catabolic fractions of AGE-modified proteins bear dicarbonyl Maillard reaction intermediates, which are a much more aggressive menace to plasma and tissue proteins as compared to the role formerly attributed to glucose . Therefore, plasma proteins can become glycated by glucose itself or by the more potent ‘second generation’ agents (if we use an analogy with antibiotics). For instance, using a rat model of diabetes, we have recently shown that AGE-peptides modify IgG. This change in IgG could lead to functional impairment of antibody molecules, and be linked to the well-known increase in susceptibility to infection seen in diabetic rat models and in man. Further studies are needed to ascertain the correctness of this hypothesis.

Our work has precisely shown that AGE-peptides are filtered by the glomerulus and catabolized in part by the endolysosomal system of the proximal convoluted tubule, as shown in the figure. Reabsorption could represent an AGE-receptor-mediated mechanism triggering several cell responses including cytokine secretion and oxidation reactions. Following this line of reasoning, one might hypothesize that in diabetes an increase in these processes could participate in the interstitial fibrosis reaction accompanying the characteristic glomerulosclerosis of end-stage renal disease. In the long run, we might speculate that the increased tubular charge of AGE-peptides due to diabetes may overwhelm the whole process and lead to tubular disorders . In a nutshell, AGE peptides increase in diabetes (excess of production) and in kidney failure (decreased excretion).

Finally, AGE-peptides also bind covalently to phospholipids and they react with membrane phospholipids if present in high local concentrations (such as shown by us in lysosomes) and if sufficient time is allowed. An accumulation of these adducts in tubular lysosomes might prove to be one further aggression to membranes and yet another process contributing to the overall toxicity.

In summary, in addition to glucose-derived AGEs, the endogenously produced degradation products, AGE peptides, can amplify tissue damage and thus act as distinct toxins. The effects may particularly accelerate the deleterious effect of glucose in certain individuals that are genetically susceptible to diabetic complications.

It is believed that circulating AGE-peptides are probably the result of incomplete catabolism of AGE-proteins by macrophages and other cells which are in their way to be excreted by the kidneys.

. In a nutshell, AGE peptides increase in diabetes (excess of production) and in kidney failure (decreased excretion).

Our work has precisely shown that AGE-peptides are filtered by the glomerulus and catabolized in part by the endolysosomal system of the proximal convoluted tubule, as shown in the figure. Reabsorption could represent an AGE-receptor-mediated mechanism triggering several cell responses including cytokine secretion and oxidation reactions.

What about stroke and heart attacks? What is the case for diabetic macroangiopathy and glycation?

Numerous questions remain unanswered with regards to the role of hyperglycemia in macrovascular complications seen in types 1 and 2 diabetics and how treatment of hyperglycemia may affect these complications. With the ability to measure HbA1c levels, the DCCT found a 41% reduction in the risk for macrovascular events, which was not statistically significant because of the low frequency of these events in that population. Nevertheless, these data certainly suggest a possible role for hyperglycemia in accelerating the atherosclerotic process in patients with type 1 diabetes. In the same line, epidemiologic analyses of UKPDS data, have shown strong associations between blood glucose control and the risk of cardiovascular disease and all-cause mortality. There was a 16% reduction (not statistically significant) in the risk of myocardial infarction and sudden death in the intensively treated group. Nonetheless, as pointed out earlier, these studies do not prove as yet that high blood glucose causes these complications and that intensive treatment to lower glucose would reduce the risk. As pointed out earlier, an interesting observation from the UKPDS is that metformin decreased the risks of diabetes-related deaths and myocardial infarction when compared with other conventional treatments.

Many of the pathways shown before for micro-apply also to macroangiopathy. Arterial wall collagen bearing AGEs can trap LDL and IgG particles, which in turn can accumulate in the intima. In this way, they would be prone to local oxidation and uptake by monocyte-macrophages. At the same time, endothelial cell activation may mediate the deposition of atheroma, since oxidized low-density lipoprotein causes endothelial cell activation . On the other hand, activation of monocyte receptors by AGEs on vascular wall proteins, such as collagen and elastin, would trigger the aforementioned sequence of cytokine-mediated inflammatory reactions.

Vascular diabetic complications may be due in part to chronic endothelial cell activation . The picture is incomplete as yet, for some mechanisms of endothelial cell activation have been observed only in vitro or in animals.

On the other hand, as the diagram below summarizes, extensive literature shows a role for the glycation of lipoproteins in atherogenesis. Early glycation of apoB, apoAs and apo E has been described , and abnormal metabolism of glycated forms of LDL and HDL have been reported . Enhanced glycation may have direct effects and may also amplify the effects of oxidative stress on lipoproteins . Thus, glycation has been shown not only to increase the susceptibility of LDL to oxidation but also, as shown earlier, to enhance the propensity of vessel wall structural proteins to bind plasma proteins, including LDL, and thus to contribute to a more marked oxidative modification of LDL. Glycated and oxidized lipoproteins induce cholesteryl ester accumulation in human macrophages and may promote platelet and endothelial cell dysfunction.

With regards to high-density lipoproteins (HDL), we have shown that in vitro activation of lecithin- cholesterol acyltransferase (LCAT) by glycated apolipoprotein A-I (apoA-I is the major apoprotein in HDL) was lower than the activation by native apolipoprotein A-I . These data were confirmed by others in diabetic patients. Because LCAT affords a driving force in reverse cholesterol transport, it is provoking to conjecture that this abnormal activation may be associated with a reduction in reverse cholesterol transport and accelerated atherosclerosis in diabetic patients.

Even if it is too early to conclude that reduction of hyperglycemia would have as great an impact on lowering macrovascular-disease risk, as it has on microvascular-disease risk, these studies afford further stimulus to explore this issue.

The many links between glycation and atherosclerosis

Extensive literature shows a role for the glycation of lipoproteins in atherogenesis

Can't we do something about it?

Therapeutic agents that inhibit AGE formation have made it possible to investigate the role of AGEs in the development of diabetic complications using animal models . The main AGE inhibitor discovered is aminoguanidine, and it has been studied in considerable detail. As shown in our figure, aminoguanidine reacts mainly with dicarbonyl intermediates such as 3-deoxyglucosone rather than with fructosamine products on proteins. It must be noted that in addition to inhibiting AGE formation, aminoguanidine inhibits the inducible form of nitric oxide synthase in vitro. In vivo, however, concentrations ten times higher than those used to inhibit AGEs are needed to change nitric oxide concentrations in a significant manner.

The effects of aminoguanidine on diabetic pathology have been investigated first in animal models. The prevention of AGE formation by aminoguanidine treatment delays the evolution of the microvascular lesions found in diabetic animals either in the retina or the glomeruli. Primary and secondary prevention with aminoguanidine has been successfully employed to ameliorate diabetic retinopathy in the rat. In some studies, treatment with aminoguanidine reduced endothelial proliferation and completely arrested pericyte dropout, but it did not completely attenuate the progression of vascular occlusion.

When renal failure was produced in streptozotocin-induced diabetic rats by surgical reduction of renal mass and aminoguanidine was administered, the treated rats had significantly superior survival than that of untreated uremic diabetic animals. The extended survival rate in this rat model of uremia and diabetic nephropathy suggests that aminoguanidine may prove beneficial in human diabetes.

Other researchers investigated the effect of aminoguanidine on slowing of motor nerve conduction velocity of the sciatic nerve in streptozocin-induced diabetic rats. Motor nerve conduction velocity was inversely correlated with AGE levels, and aminoguanidine improved nerve conduction probably through decreasing the AGE level in the peripheral tissues . It may have a therapeutic potential in controlling diabetic peripheral neuropathy.

Will AGE inhibitors also prevent diabetic complications in humans? At what point in the natural history of the disease would treatment be most effective? As the intimate sequence of the steps that lead to vascular diseases associated with diabetes per se are yet not fully understood, we have no precise answer to these questions.

Aminoguanidine, very potent but..

Longer trials in better-defined populations are needed before the effectiveness of these inhibitors can be proven. Some problems of toxicity have been encountered in a phase III clinical trial with aminoguanidine, so this drug should be considered a prototype for many new molecules which are being synthesized and tried in vitro at present.

When safe anti-glycation drugs are marketed we will be attacking the root of the problem instead of treating the end-stage resulting pathology.

Not only outside the cells....

It has been recently demonstrated that AGEs also form on cell proteins in vivo . More so, they also form on DNA in vitro. If AGEs also form on DNA in vivo, deleterious effects on gene expression may occur and intracellular AGE formation on cell proteins may thus, in turn, affect DNA function. Actually, the extremely rapid rate of AGE formation, shown by us, on liver histones points in this direction. Histones from the liver of rats after only one month of hyperglycemia showed AGE levels three-fold higher than those of their age-matched controls, and accumulation of AGEs on histones increased with the duration of the disease . This suggests a possible role for intracellular glycation in the increased teratogeny associated with diabetes mellitus .

Glycation inside our cells may be even faster since small metabolites are much more reactive than sugars

2 of them, methylglyoxal and glyceraldehyde are focus of many research projects